Re: [etch] driver updates for 2.6.18?
On Thu, 23 Aug 2007, dann frazier wrote:
> On Thu, Aug 23, 2007 at 01:34:26PM +0200, Bastian Blank wrote:
> > seems to work fine with the xen in etch (better than the released
> > version from Fedora). The summary of the functional changes is
> > | 244 files changed, 11100 insertions(+), 6047 deletions(-)
> > The complete one
> > | 351 files changed, 18610 insertions(+), 7485 deletions(-)
> > Most of the changes are different ways of porting the same functionalty
> > to the version.
> >
> > The following problems should be fixed:
> > - Event channel code is oopsy on smp systems.
> > - PCI backend is broken.
> > - Network backend looses all running connections after some months (this
> > may be also a hypervisor problem).
> > - Incorrect license specifications in the xen headers.
>
> Each of those sounds like a reasonable thing to fix in a point
> release, but such a large change does not. I don't see how we can
> demonstrate confidence that such a change will not cause regressions.
xen in etch is way beyond broken.
i don't even start looking when the bug report mentions xen.
the bts has *way* too much xen bugs.
i doubt it is worth the effort to isolate them to some specific
changesets.
please take the bts fact in consideration for the etch update xen decision.
--
maks
Reply to: